JPS61266964A - Processor for period data on pulse signal - Google Patents

Abstract

PURPOSE:To perform arithmetic processing only with correct period data by deciding and masking error data of input period data which is not within a reference range and preventing the error data from being read in when a noise pulse is generated. CONSTITUTION:When a rotor 1 makes one turn, a rotary sensor 2 generates a sine wave of frequency proportional to the rotation and inputs it to a waveform shaping circuit 3 and a counter 4 measures the period of the waveform shaping output with a reference clock. A microcomputer 5 reads the period data and compares it with a specific reference value to decide that the period data is error data unless the data is within the range determined by the reference value. The error period data and period data inputted right before and after the period data are masked to compute the standard deviation and arithmetic mean value based upon only correct period data, thereby computing the rotating speed of wheels in a specific run section.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an apparatus for processing periodic data of a plurality of pulse signals generated sequentially over time. The present invention can be applied to a device that calculates the rotational speed based on periodic data of a pulse signal output from a rotation sensor that detects the rotational speed of a wheel.

[Conventional technology]

2. Description of the Related Art Conventionally, a technique is known in which a rotation sensor using an electromagnetic pick amplifier is used to output a pulse signal corresponding to the wheel speed, and the wheel speed is measured by calculating the period of this pulse waveform.

In this case, the output signal from the electromagnetic pickup has a sine waveform as shown in waveform S in FIG. 6, and is shaped into a rectangular pulse waveform as shown in waveform S2 by waveform processing. Therefore, the period Tl of this pulse waveform S2,
The rotational speed is determined by calculating T2, . . . T6.

[Problem that the invention seeks to solve]

Here, when noise shown by Sz occurs in the waveform Sl, a pulse SZ+ is generated in the waveform S2 in response. The period T of this pulse SZ+ is extremely short compared to the periods of other pulse signals (for example, Ts, T+),
It can be seen from the nature of the pulse waveform of the rotation signal that it is clearly incorrect data.

Furthermore, if this noise pulse S21 is generated at a very eccentric position as shown in FIG.
It is difficult to determine whether K+ is noise generated in the middle of pulse StO or noise generated in the middle of pulse So+. This noise pulse Sil+ is the pulse S.

If it occurs in the middle of , not only the noise pulse SZ+ but also the pulse 52m becomes erroneous data.

The technical problem of the present invention is to perform arithmetic processing using only correct periodic data without reading erroneous data when a noise pulse occurs.

[Means for Solving the Problems] Therefore, as shown in FIG. 1, the present invention comprises: an input means M for inputting periodic data of a plurality of pulse signals generated sequentially over time; a determining means M2 that compares the periodic data with a predetermined reference value and determines the periodic data as incorrect data when the periodic data is not within the range determined by the reference value; and the period determined as the incorrect data. a masking means M for masking data and periodic data input before and after the periodic data; and a masking means M for reading only the data not masked by the masking means among the periodic data inputted by the inputting means. A technical means is adopted in which a calculation means M4 for performing calculation processing is provided.

[Effect]

By adopting the above technical means, it is possible to identify erroneous data that is not within the range defined by the reference value among multiple period data input in chronological order, and to identify this erroneous data and the periods input before and after this erroneous data. Since data is also masked, only correct periodic data is selected and processed. For example, in the waveform diagram 82 shown in FIG.
Not only the + period data T3 but also the period data TM of the pulse S2° and the period data T4 of the pulse S are masked.

〔Example〕

The present invention will be described in detail below based on embodiments shown in the drawings. FIG. 2 shows a first embodiment in which the present invention is applied to a device for detecting the rotational speed of a wheel. In FIG. 2, a rotor 1, which is a rotating body provided integrally with the wheels of an automobile, includes:
A groove 36 per rotation is formed, and the tip of a rotation sensor 2 using a well-known electromagnetic pickup approaches the groove.

When the rotor 1 rotates, the rotation sensor 2 generates a sine wave with a frequency proportional to the rotation. This sine wave is input to the waveform shaping circuit, and the waveform is shaped by the comparator 3b for eliminating mutual influence by the buffer 3a, and the waveform is as follows.
Time T from one rise to the next rise (T=T
, , 'rz, . . . ) are measured by a counter 4 that measures the period using a reference clock. Then, the microcomputer 5 reads the data of the length of one cycle. The microcomputer masks erroneous cycle data from these input cycle data, reliably recognizes only the correct cycle data, and calculates the standard deviation of the correct cycle data,
Statistical processing such as calculating an arithmetic mean value is performed to calculate the rotational speed of the wheels in a predetermined travel section.

Therefore, the microcomputer 5 has a built-in control program in its storage device that determines whether the above-mentioned periodic data is correct or incorrect and masks erroneous data.This control program will be explained in detail below. FIG. 3 shows a flowchart of this control program, and the control program will be explained below along with this flowchart. step 100
initializes the memory, and then proceeds to step 101.
Proceed to. Step 101 is a step of inputting the current cycle data T measured by the counter 4, and then the process proceeds to step 102. Step 102 is step 2 of checking FLAG, and if FLAG=1, step 112
The process proceeds to step 101 and is set to 0. If FLAG=0, the process advances to the next step 103. Step 103 is a step of checking the presence or absence of the preceding cycle data MEA.
If EA=0, proceed to step 109 and perform step 1
The current cycle data T input in step 01 is stored in the memory MEA, and the process returns to step 101. If MEA~0, proceed to the next step 104.

Step 104 is a step of checking the presence or absence of preceding cycle data MEB, and if MEB=O, step 1
Proceed to step 13. In step 113, the current cycle data and the previous cycle data MEA are compared, and if MEA/2≧T, nothing is done and the process returns to step 101. If MEA/2<T, the process proceeds to step 114, stores the current cycle data T in the memory EMB, and returns to step 101. Step 1
If MEB~O in step 04, the process advances to the next step 105.

Step 105 compares the current cycle data T and the preceding cycle data MEA, and if MEA/2≧T, nothing is done and the process returns to step 101. If MEA/2<T, the process proceeds to step 114, stores the current cycle data T in the memory MEB, and returns to step 101.

If MEA/2<T in step 105, the process proceeds to the next step 106. In step 106, the memory M
Read the periodic data of EA into the RAM as correct periodic data.

Steps 107 and 108 are steps for updating data, in which the preceding cycle data MEB is moved to the preceding cycle data MEA, the current cycle data T is moved to the preceding cycle data MEB, and the process returns to step 101.

Next, the overall flow of this control program will be explained. If no single noise occurs, first step 100→
101→102→103→109→101→102→1
03→104→113→114→101-4102→1
Read the smallest data MEA in the order of 03 → 104 → 105 → 106-107 → 108, and if no single noise occurs after that, step 101 → 102-103
→ 104 → 105 → 106 → 107 → 108 → 101 closed loop is repeatedly executed.

If a single noise occurs during this process, step 101 →
102 → 103 → 104 → 105 → 110 → 111 → 1
01→102→112→101→102→103→10
4→113→114→101→102→103→104
Processing is performed in the order of →105 →106 →107-108 to ensure masking of erroneous cycle data due to single noise.

If a single noise occurs continuously during the process, step 101 → 102 → 103 → 104 → 105 → 110
→１１１→１０１→１０２→１１２→１０１→１０２→
Processing is performed in the order of 103 → 104 → 113, and then steps 101 → 102 → 103 → 104 → 113 → 10
1 closed loop is repeatedly executed for the number of continuous single-shot noises to ensure masking of data having an erroneous period due to continuous single-shot noises. When the single-shot noise disappears, steps 101→102→1 are performed in the same way as above.
A closed loop of 03→104→105→106→107→108 is repeatedly executed to input data.

Next, a second embodiment of the present invention will be described.

In the second embodiment, focusing on the fact that in the case of single-shot noise, the time from its rise (fall) to its fall (rise) is extremely short (usually several μsec or less), and the time until
Measures both the time from one falling edge to the next rising edge, and if each time is extremely short compared to a certain reference value, it is determined to be single-shot noise, and false data is reliably masked to ensure correct data. By reading in the data and adding two consecutive pieces of data, it is possible to reliably calculate and recognize the correct periodic data. When adding two consecutive pieces of data, add the time from one rising edge to the next falling edge and the time from one falling edge to the next rising edge to obtain periodic data. , it will be period data from one rising edge to the next rising edge, and if you do the reverse, it will be period data from one falling edge to the next falling edge. Which period data to select can be easily resolved during calculation. There is also the advantage of being able to do so.

FIG. 4 shows an example in which the second embodiment is applied to wheel rotational speed detection in the same way as the first embodiment. As shown in FIG. 4, in the second embodiment, the waveform shaped by the waveform shaping circuit 3 is 1, and the time from rise to fall is T A (T a r , T a z ).

...) with a reference clock, and the time T from a falling edge to the next rising edge.

(Tel, 'r, . . . ) are measured separately with a counter 46 that measures them using a reference clock. These data are read in alternately by the microcomputer 5.

The other configurations except for the microcomputer 5 are as follows.
Since it is the same as the first embodiment shown in FIG. 2, the explanation will be omitted, and the control program for masking the erroneous cycle data of the microcomputer 5 will be explained in detail.

FIG. 5 is a flowchart showing the processing procedure of the control program, and the following will be explained according to this flowchart.

Step 200 is a step for initializing the memory, and the process then proceeds to step 201. Step 201 repeats step 201 if the first data is TA or T. In the case of TA, the process advances to the next step 202. In steps 202 to 209, data T
A, T! 1 is stored in four memories M1 to M4. (In other words, data for two cycles is stored in memory). After this,
Proceeding to step 21Q, step 210 includes memory M1
If Ml<H, the process advances to step 217 and checks M4.

In step 217, if M4≧H, step 202
Return to If M4<H, proceed to step 218 and mask 2 TA, T, which will be input next, and step 2
Return to 02. If M1≧H in step 210, the process advances to the next step 211. Step 211 is step 21
0 is basically the same, so M2 is checked and the process proceeds to step 212. Step 212 is also basically the same as step 210, and checks M3. Step 21
When step 2 is completed, the process proceeds to step 213, where correct period data T is calculated. Here, the calculation formula is T=M1+M2
．． , this periodic data T is read into the storage device in step 214 as correct data. Thereafter, the process proceeds to step 2ts, it6, and the memory is updated.

Therefore, when this control program is executed, if no single noise occurs, step 200→201
→202→203→204→205→206→207→
208→209→210→211→212→213→2
If processing is performed in the order of 14→215→216, steps 216-206→207-208→209→210→
A closed loop of 211-212→213→214→215 is repeatedly executed.

Step 2: Check whether single-shot noise occurred during this process.
10 to 212, and if it is determined to be single-shot noise, the last data of the four pieces of data is checked in step 217, and if this data is also determined to be single-shot noise,
The four data sets M1 to M4 and the next two data sets are masked and the process returns to step 202. If the last data is not determined to be single-shot noise, only the M1 to M404 data are masked and the process returns to step 202. By repeating this calculation, erroneous data can be masked reliably.

〔Effect of the invention〕

As described above, according to the present invention, erroneous data and data suspected of being erroneous data are masked, and only correct cyclic data is used to perform arithmetic processing on periodic data (for example, calculation of the average value of periodic data). This has the effect of eliminating the influence of noise and increasing the reliability of arithmetic processing.

[Brief explanation of the drawing]

FIG. 1 is an overall configuration diagram of the present invention, FIG. 2 is a configuration diagram of a first embodiment of the present invention, FIG. 3 is a flowchart of the control program of the microcomputer shown in FIG. 2, and FIG. 4 is a diagram of the present invention. FIG. 5 is a flowchart of the control program of the microcomputer shown in FIG. 4, and FIG. 6 is a waveform diagram showing an example of a pulse signal waveform. 2... Rotation sensor, 3... Waveform shaping circuit, 4...
Counter, 5...microcomputer. Representative Patent Attorney Takashi Okabe Fig. 1 Fig. 2 WJ Fig. 3 Fig. 4 D Fig. 6

Claims (1)

[Scope of Claims] Input means for inputting periodic data of a plurality of pulse signals generated sequentially over time, and comparing the inputted periodic data with a predetermined reference value, and comparing the inputted periodic data with a predetermined reference value, and determining whether certain periodic data is equal to the reference value. determining means for determining the periodic data as incorrect data when it is not within a range determined by the above; and masking means for masking the periodic data determined to be the incorrect data and the periodic data input before and after the periodic data. A pulse signal periodic data processing device characterized by comprising: a calculation means for reading periodic data not masked by the masking means from among the periodic data inputted by the inputting means and performing calculation processing. .